Action redirection method, apparatus, and computing device cluster

Abstract

An action redirection method includes: segmenting a target character based on semantics of different parts of the target character, to obtain a plurality of first body segments related to the target character; modeling each first body segment based on a first coordinate system marked on the first body segment, to obtain a plurality of first geometric models, where one first geometric model represents a shape of one body segment; performing collision detection on the plurality of first geometric models; and when first geometric models collide with each other, adjusting a distance between the first geometric models that collide with each other, and adjusting an action sequence of the target character based on an adjusted distance. Thus, a coordinate system is constructed for a body segment of the target character, and a geometric model is output through modeling, so that accurate geometric expressing of the body segment is implemented.

1 . An action redirection method, comprising:
 segmenting a target character based on semantics of different parts of the target character, to obtain a plurality of first body segments related to the target character;   modeling each first body segment based on a first coordinate system marked on the first body segment, to obtain a plurality of first geometric models, wherein one first geometric model represents a shape of one body segment;   performing collision detection on the plurality of first geometric models; and   when the first geometric models collide with each other, adjusting a distance between the first geometric models that collide with each other, and adjusting an action sequence of the target character based on an adjusted distance, wherein the action sequence of the target character is obtained by transferring an action sequence of a source character.   
     
     
         2 . The method according to  claim 1 , wherein modeling each first body segment based on the first coordinate system marked on the first body segment comprises:
 determining, based on the first coordinate system marked on the first body segment, a coordinate expression of a point on a surface of the first body segment in the first coordinate system; and   training a to-be-trained model based on the coordinate expression, to obtain the first geometric model.   
     
     
         3 . The method according to  claim 1 , wherein performing collision detection on the plurality of first geometric models comprises:
 for any two first geometric models, determining a minimum distance between the 1 st  geometric model and the 2 nd  geometric model, wherein the minimum distance is a distance between a first point inside the 1 st  geometric model and a second point inside the 2nd geometric model;   inputting at least a part of a coordinate expression of the first point in a coordinate system associated with the 2 nd  geometric model into the 2 nd  geometric model, to obtain a first distance, and inputting at least a part of a coordinate expression of the second point in a coordinate system associated with the 1 st  geometric model into the 1 st  geometric model, to obtain a second distance; and   when a sum of the first distance and the second distance is greater than the minimum distance, determining that the 1 st  geometric model collides with the 2 nd  geometric model.   
     
     
         4 . The method according to  claim 1 , wherein before performing collision detection on the plurality of first geometric models, the method further comprises:
 segmenting the source character based on semantics of different parts of the source character, to obtain a plurality of second body segments related to the source character;   modeling, based on the first coordinate system and first key points marked on all the first body segments and second coordinate system and second key points marked on all the second body segments, mapping relationships between the first body segments and the second body segments associated with same parts of the target character and the source character, to obtain a plurality of mapping models, wherein one mapping model is used to express a mapping relationship between one first body segment and one second body segment;   determining a location relationship between a second body part of the target character and each first body segment based on the plurality of mapping models and a location relationship between a first body part of the source character and each second body segment, wherein the first body part is the same as the second body part;   determining a final location of the second body part in a world coordinate system based on the location relationship between the second body part of the target character and each first body segment; and   adjusting the action sequence of the target character based on the final location, so that the second body part is located at the final location.   
     
     
         5 . The method according to  claim 4 , wherein both the first coordinate system and the second coordinate system are cylindrical coordinate systems;
 a first location relationship between the first body part and the second body segment comprises:
 a first location of a first projection point of the first body part and on a central axis of the first coordinate system, 
 a first direction relationship between a direction from the first projection point to the first body part and an axis except for the central axis, a second direction relationship between the direction from the first projection point to the first body part and another axis except for the central axis, and 
 a first distance between the first projection point and the first body part; and 
   determining the location relationship between the second body part of the target character and each first body segment based on the plurality of mapping models and the location relationship between the first body part of the source character and each second body segment comprises:   inputting the first location into a mapping model related to the first location relationship, to obtain a second location of a second projection point of the second body part and on a central axis of the second coordinate system;   inputting the second location, the first direction relationship, and the second direction relationship into a first geometric model of a first body segment related to the first location relationship, to obtain a second distance between the second projection point and the second body part;   inputting the first location, the first direction relationship, and the second direction relationship into a second geometric model of the second body segment related to the first location relationship, to obtain a third distance between the first projection point and the first body part, wherein the second geometric model is obtained through modeling based on a coordinate system marked on the second body segment related to the first location relationship; and   obtaining a fourth distance based on the first distance, the second distance, and the third distance, wherein   a second location relationship between the second body part and the first body segment related to the first location relationship comprises the second location, the first direction relationship, the second direction relationship, and the fourth distance.   
     
     
         6 . The method according to  claim 4 , wherein both the first coordinate system and the second coordinate system are spherical coordinate systems;
 a first location relationship between the first body part and the second body segment comprises:
 a first relationship between a direction from the first body part to an origin of the first coordinate system and a direction of a z axis of the first coordinate system, 
 a second relationship between the direction from the first body part to the origin of the first coordinate system and a direction of an x axis of the first coordinate system, 
 a third relationship between the direction from the first body part to the origin of the first coordinate system and a direction of a y axis of the first coordinate system, and 
 a first distance from the first body part to the origin of the first coordinate system; and 
   determining the location relationship between the second body part of the target character and each first body segment based on the plurality of mapping models and the location relationship between the first body part of the source character and each second body segment comprises:   inputting the first relationship into a mapping model related to the first location relationship, to obtain a fourth relationship between a direction from the second body part to the origin of the first coordinate system and a direction of a z axis of the second coordinate system;   inputting the fourth relationship, the second relationship, and the third relationship into a first geometric model of a first body segment related to the first location relationship, to obtain a second distance between the second body part and an origin of the second coordinate system;   inputting the first relationship, the second relationship, and the third relationship into a second geometric model of the second body segment related to the first location relationship, to obtain a third distance between the first body part and the origin of the first coordinate system, wherein the second geometric model is obtained through modeling based on a coordinate system marked on the second body segment related to the first location relationship; and   obtaining a fourth distance based on the first distance, the second distance, and the third distance, wherein   a second location relationship between the second body part and the first body segment related to the first location relationship comprises the fourth relationship, the second relationship, the third relationship, and the fourth distance.   
     
     
         7 . The method according to  claim 4 , wherein the first body segment is associated with a first part of the target character, and the second body segment is associated with a second part of the source character; and
 when the first part is the same as the second part, a type of a coordinate system marked on the first body segment is the same as a type of the coordinate system marked on the second body segment, and a type of the first key point marked on the first body segment is the same as a type of the second key point marked on the second body segment.   
     
     
         8 . An action redirection device, comprising:
 a processor, and   a memory coupled to the processor to store instructions, which when executed by the processor, cause the action redirection device to:   segment a target character based on semantics of different parts of the target character, to obtain a plurality of first body segments related to the target character;   model each first body segment based on a first coordinate system marked on the first body segment, to obtain a plurality of first geometric models, wherein one first geometric model represents a shape of one body segment; and   perform collision detection on the plurality of first geometric models, wherein   when the first geometric models collide with each other, adjust a distance between the first geometric models that collide with each other, and adjust an action sequence of the target character based on an adjusted distance, wherein the action sequence of the target character is obtained by transferring an action sequence of a source character.   
     
     
         9 . The device according to  claim 8 , wherein when modeling each first body segment based on the first coordinate system marked on the first body segment, the instructions, when executed, further cause the device to:
 determine, based on the first coordinate system marked on the first body segment, a coordinate expression of a point on a surface of the first body segment in the first coordinate system; and   train a to-be-trained model based on the coordinate expression, to obtain the first geometric model.   
     
     
         10 . The device according to  claim 9 , wherein when performing collision detection on the plurality of first geometric models, the instructions, when executed, further cause the device to:
 for any two first geometric models, determine a minimum distance between the 1 st  geometric model and the 2 nd  geometric model, wherein the minimum distance is a distance between a first point inside the 1 st  geometric model and a second point inside the 2 nd  geometric model;   input at least a part of a coordinate expression of the first point in a coordinate system associated with the 2 nd  geometric model into the 2 nd  geometric model, to obtain a first distance, and input at least a part of a coordinate expression of the second point in a coordinate system associated with the 1 st  geometric model into the 1 st  geometric model, to obtain a second distance; and   when a sum of the first distance and the second distance is greater than the minimum distance, determine that the 1 st  geometric model collides with the 2 nd  geometric model.   
     
     
         11 . The device according to  claim 9 , wherein before performing collision detection on the plurality of first geometric models, the instructions, when executed, further cause the device to:
 segment the source character based on semantics of different parts of the source character, to obtain a plurality of second body segments related to the source character;   model, based on the first coordinate system and first key points marked on all the first body segment and second coordinate system and second key points marked on all the second body segments, mapping relationships between the first body segments and the second body segments associated with same parts of the target character and the source character, to obtain a plurality of mapping models, wherein one mapping model is used to express a mapping relationship between one first body segment and one second body segment;   determine a location relationship between a second body part of the target character and each first body segment based on the plurality of mapping models and a location relationship between a first body part of the source character and each second body segment, wherein the first body part is the same as the second body part;   determine a final location of the second body part in a world coordinate system based on the location relationship between the second body part of the target character and each first body segment; and   adjust the action sequence of the target character based on the final location, so that the second body part is located at the final location.   
     
     
         12 . The device according to  claim 11 , wherein both the first coordinate system and the second coordinate system are cylindrical coordinate systems;
 a first location relationship between the first body part and the second body segment comprises:
 a first location of a first projection point of the first body part and on a central axis of the first coordinate system, 
 a first direction relationship between a direction from the first projection point to the first body part and an axis except for the central axis, 
 a second direction relationship between the direction from the first projection point to the first body part and another axis except for the central axis, and 
   a first distance between the first projection point and the first body part; and   when determining the location relationship between the second body part of the target character and each first body segment based on the plurality of mapping models and the location relationship between the first body part of the source character and each second body segment, the instructions, when executed, further cause the device to:   input the first location into a mapping model related to the first location relationship, to obtain a second location of a second projection point of the second body part and on a central axis of the second coordinate system;   input the second location, the first direction relationship, and the second direction relationship into a first geometric model of a first body segment related to the first location relationship, to obtain a second distance between the second projection point and the second body part;   input the first location, the first direction relationship, and the second direction relationship into a second geometric model of the second body segment related to the first location relationship, to obtain a third distance between the first projection point and the first body part, wherein the second geometric model is obtained through modeling based on a coordinate system marked on the second body segment related to the first location relationship; and   obtain a fourth distance based on the first distance, the second distance, and the third distance, wherein   a second location relationship between the second body part and the first body segment related to the first location relationship comprises the second location, the first direction relationship, the second direction relationship, and the fourth distance.   
     
     
         13 . The device according to  claim 11 , wherein both the first coordinate system and the second coordinate system are spherical coordinate systems;
 a first location relationship between the first body part and the second body segment comprises:
 a first relationship between a direction from the first body part to an origin of the first coordinate system and a direction of a z axis of the first coordinate system, 
 a second relationship between the direction from the first body part to the origin of the first coordinate system and a direction of an x axis of the first coordinate system, 
 a third relationship between the direction from the first body part to the origin of the first coordinate system and a direction of a y axis of the first coordinate system, and 
 a first distance from the first body part to the origin of the first coordinate system; and 
   when determining the location relationship between the second body part of the target character and each first body segment based on the plurality of mapping models and the location relationship between the first body part of the source character and each second body segment, the instructions, when executed, further cause the device to:   input the first relationship into a mapping model related to the first location relationship, to obtain a fourth relationship between a direction from the second body part to the origin of the first coordinate system and a direction of a z axis of the second coordinate system;   input the fourth relationship, the second relationship, and the third relationship into a first geometric model of a first body segment related to the first location relationship, to obtain a second distance between the second body part and an origin of the second coordinate system;   input the first relationship, the second relationship, and the third relationship into a second geometric model of the second body segment related to the first location relationship, to obtain a third distance between the first body part and the origin of the first coordinate system, wherein the second geometric model is obtained through modeling based on a coordinate system marked on the second body segment related to the first location relationship; and   obtain a fourth distance based on the first distance, the second distance, and the third distance, wherein   a second location relationship between the second body part and the first body segment related to the first location relationship comprises the fourth relationship, the second relationship, the third relationship, and the fourth distance.   
     
     
         14 . The device according to  claim 11 , wherein the first body segment is associated with a first part of the target character, and the second body segment is associated with a second part of the source character; and
 when the first part is the same as the second part, a type of a coordinate system marked on the first body segment is the same as a type of the coordinate system marked on the second body segment, and a type of the first key point marked on the first body segment is the same as a type of the second key point marked on the second body segment.   
     
     
         15 . A non-transitory computer-readable storage medium having instructions stored therein, which when executed by a processor, cause the processor to:
 segment a target character based on semantics of different parts of the target character, to obtain a plurality of first body segments related to the target character;   model each first body segment based on a first coordinate system marked on the first body segment, to obtain a plurality of first geometric models, wherein one first geometric model represents a shape of one body segment;   perform collision detection on the plurality of first geometric models; and   when the first geometric models collide with each other, adjusting a distance between the first geometric models that collide with each other, and adjusting an action sequence of the target character based on an adjusted distance, wherein the action sequence of the target character is obtained by transferring an action sequence of a source character.   
     
     
         16 . The non-transitory computer-readable storage medium according to  claim 15 , wherein to model each first body segment based on the first coordinate system marked on the first body segment, the instructions, when executed, further cause the processor to:
 determine, based on the first coordinate system marked on the first body segment, a coordinate expression of a point on a surface of the first body segment in the first coordinate system; and   train a to-be-trained model based on the coordinate expression, to obtain the first geometric model.   
     
     
         17 . The non-transitory computer-readable storage medium according to  claim 15 , wherein to perform collision detection on the plurality of first geometric models, the instructions, when executed, further cause the processor to:
 for any two first geometric models, determine a minimum distance between the 1 st  geometric model and the 2 nd  geometric model, wherein the minimum distance is a distance between a first point inside the 1 st  geometric model and a second point inside the 2 nd  geometric model;   input at least a part of a coordinate expression of the first point in a coordinate system associated with the 2 nd  geometric model into the 2 nd  geometric model, to obtain a first distance, and inputting at least a part of a coordinate expression of the second point in a coordinate system associated with the 1 st  geometric model into the 1 st  geometric model, to obtain a second distance; and   when a sum of the first distance and the second distance is greater than the minimum distance, determine that the 1 st  geometric model collides with the 2 nd  geometric model.   
     
     
         18 . The non-transitory computer-readable storage medium according to  claim 15 , wherein before performing collision detection on the plurality of first geometric models, the instructions, when executed, further cause the processor to:
 segment the source character based on semantics of different parts of the source character, to obtain a plurality of second body segments related to the source character;   model, based on the first coordinate system and first key points marked on all the first body segments and second coordinate system and second key points marked on all the second body segments, mapping relationships between the first body segments and the second body segments associated with same parts of the target character and the source character, to obtain a plurality of mapping models, wherein one mapping model is used to express a mapping relationship between one first body segment and one second body segment;   determine a location relationship between a second body part of the target character and each first body segment based on the plurality of mapping models and a location relationship between a first body part of the source character and each second body segment, wherein the first body part is the same as the second body part;   determine a final location of the second body part in a world coordinate system based on the location relationship between the second body part of the target character and each first body segment; and   adjust the action sequence of the target character based on the final location, so that the second body part is located at the final location.   
     
     
         19 . The non-transitory computer-readable storage medium according to  claim 18 , wherein both the first coordinate system and the second coordinate system are cylindrical coordinate systems;
 a first location relationship between the first body part and the second body segment comprises:
 a first location of a first projection point of the first body part and on a central axis of the first coordinate system, 
 a first direction relationship between a direction from the first projection point to the first body part and an axis except for the central axis, 
 a second direction relationship between the direction from the first projection point to the first body part and another axis except for the central axis, and 
 a first distance between the first projection point and the first body part; and 
   to determine the location relationship between the second body part of the target character and each first body segment based on the plurality of mapping models and the location relationship between the first body part of the source character and each second body segment, the instructions, when executed, further cause the processor to:   input the first location into a mapping model related to the first location relationship, to obtain a second location of a second projection point of the second body part and on a central axis of the second coordinate system;   input the second location, the first direction relationship, and the second direction relationship into a first geometric model of a first body segment related to the first location relationship, to obtain a second distance between the second projection point and the second body part;   input the first location, the first direction relationship, and the second direction relationship into a second geometric model of the second body segment related to the first location relationship, to obtain a third distance between the first projection point and the first body part, wherein the second geometric model is obtained through modeling based on a coordinate system marked on the second body segment related to the first location relationship; and   obtain a fourth distance based on the first distance, the second distance, and the third distance, wherein   a second location relationship between the second body part and the first body segment related to the first location relationship comprises the second location, the first direction relationship, the second direction relationship, and the fourth distance.   
     
     
         20 . The non-transitory computer-readable storage medium according to  claim 18 , wherein both the first coordinate system and the second coordinate system are spherical coordinate systems;
 a first location relationship between the first body part and the second body segment comprises:
 a first relationship between a direction from the first body part to an origin of the first coordinate system and a direction of a z axis of the first coordinate system, 
 a second relationship between the direction from the first body part to the origin of the first coordinate system and a direction of an x axis of the first coordinate system, 
 a third relationship between the direction from the first body part to the origin of the first coordinate system and a direction of a y axis of the first coordinate system, and 
 a first distance from the first body part to the origin of the first coordinate system; and 
   to determine the location relationship between the second body part of the target character and each first body segment based on the plurality of mapping models and the location relationship between the first body part of the source character and each second body segment, the instructions, when executed, further cause the processor to:   input the first relationship into a mapping model related to the first location relationship, to obtain a fourth relationship between a direction from the second body part to the origin of the first coordinate system and a direction of a z axis of the second coordinate system;   input the fourth relationship, the second relationship, and the third relationship into a first geometric model of a first body segment related to the first location relationship, to obtain a second distance between the second body part and an origin of the second coordinate system;   input the first relationship, the second relationship, and the third relationship into a second geometric model of the second body segment related to the first location relationship, to obtain a third distance between the first body part and the origin of the first coordinate system, wherein the second geometric model is obtained through modeling based on a coordinate system marked on the second body segment related to the first location relationship; and   obtain a fourth distance based on the first distance, the second distance, and the third distance, wherein   a second location relationship between the second body part and the first body segment related to the first location relationship comprises the fourth relationship, the second relationship, the third relationship, and the fourth distance.